ASU CoRAL lab brings 11 papers to EMNLP 2026 spanning reasoning, LLM agents and trustworthy AI
keviv9 · x · 2026-10-24
Arizona State University's CoRAL lab (PI: Vivek Gupta), with collaborators from UKPLab, Mayo Clinic, CogInt and Adobe, will present 11 works at EMNLP 2026 in Budapest:
- Reasoning & tables: CORE-T (coherent retrieval for multi-table text-to-SQL), QUIETI (query-independent table transformation), CAPTR (caption-based context pruning for multimodal tabular reasoning), Q-STR (text-to-table over cricket narratives).
- Agents & safety: CAST (critique-aware supervision for long-horizon tool-calling agents), DeALOG (decentralized log-mediated multi-agent reasoning), ChartAttack (vulnerability of LLM chart generation to malicious prompts), SentinelEdge (on-device streaming scam-call detection, demo).
- Also: SentryLine (evidence-grounded QA over evolving oncology docs, with Mayo Clinic), EviNorm (evidence-guided schema normalization for temporal tabular reasoning), and a systematic survey of trajectory feedback in diffusion LMs.
The author also co-organizes the DocInsights workshop at the same conference.
More from Research
- BeNice! method shares KV-cache across loops, scaling looped LLM depth without extra memory — stochasticchasm · 2026-10-24
- 215-state Turing machine proven to halt iff ZFC is inconsistent, verified in Metamath Zero — teortaxesTex · 2026-10-24
- OmniRay from JHU hits SOTA camera-agnostic monocular 3D reconstruction from one image — kwangmoo_yi · 2026-10-24
- AWS multi-turn RL guide: $339 lifts gpt-oss-20b Wordle solve rate 61%→84% — AWS ML Blog · 2026-10-24
- Eric Horvitz spotlights Stanford's tool-making research philosophy behind optogenetics — erichorvitz · 2026-10-24
- DreamCan does sparse visual planning for long-horizon robot manipulation — chris_j_paxton · 2026-10-24